Bank and sovereign risk feedback loops This paper studies the link between sovereign risks and the fragility of the banking sector pointing to the challenges of bank rescue operations for the state. Aitor Erce Disclaimer This Working Paper should not be reported as representing the views of the ESM. The views expressed in this Working Paper are those of the author(s) and do not necessarily represent those of the ESM or ESM policy. Working Paper Series | 1 | 2015
Transcript
Bank and sovereign risk feedback loops
This paper studies the link between sovereign risks and the fragility of the banking sector pointing to the challenges of bank rescue operations for the state.
Aitor Erce
DisclaimerThis Working Paper should not be reported as representing the views of the ESM.The views expressed in this Working Paper are those of the author(s) and do notnecessarily represent those of the ESM or ESM policy.
Working Paper Series | 1 | 2015
Working Paper Series | 1 | 2015
DisclaimerThis Working Paper should not be reported as representing the views of the ESM. The views expressed in this Working Paper are those of the author(s) and do not necessarily represent those of the ESM or ESM policy.No responsibility or liability is accepted by the ESM in relation to the accuracy or completeness of the information, including any data sets, presented in this Working Paper.
DOI 10.2852/501718EU catalogue number DW-AB-15-001-EN-N
ISSN 443-5503ISBN 978-92-95085-06-0
Aitor Erce1
Bank and sovereign risk feedback loops
I thank Antonello D’Agostino, Gong Cheng, Jon Frost, Patricia Gomez, Carlos Martins, Tomasz Orpiszewski, Cheng PG-Yan, Chander Ramaswamy, Juan Rojas, Karol Siskind and seminar participants at the European Stability Mechanism and the 2014 Symposium of Economic Analysis for their suggestions, and Sarai Criado, Gabi Perez-Quiros and Adrian Van Rixtel for sharing their CDS data. Assunta Di Chiara provided outstanding research assistance.
1 European Stability Mechanism. The author can be contacted at: [email protected].
Key words Sovereign Risk, Bank Risk, Feedback Loops, Balance Sheet Exposure, Leverage
JEL codesE58, G21, G28, H63
AbstractMeasures of sovereign and bank risk show occasional bouts of increased correlation, setting the stage for vicious and virtuous feedback loops. This paper models the macroeconomic phenomena underlying such bouts using CDS data for 10 euro area countries. The results show that sovereign risk feeds back into bank risk more strongly than vice versa. Countries with sovereigns that are more indebted or where banks have a larger exposure to their own sovereign, suffer larger feedback loop effects from sovereign risk into bank risk. In the opposite direction, in countries where banks fund their activities with more foreign credit and support larger levels of non-performing loans, the feedback from bank risk into sovereign risk is stronger. According to model estimates, financial rescue operations can increase feedback effects from bank risk into sovereign risk. These results can be useful for the official sector when deciding on the form of financial rescues.
Introduction
As the global crisis engulfed a number of economies into a perverse spiral of �s-cal and �nancial distress, the interconnectedness between banks and sovereignshas attracted increasing attention. On the one hand, a number of countriesfaced severe banking crises, whose management contributed to the subsequent�scal crisis. Arguably, this is what happened to Iceland, where the materializa-tion of contingent claims brought havoc onto the sovereign�s balance sheet.1 Onthe other hand, pro-cyclical �scal policy and a lack of competitiveness led to asovereign debt crisis in Greece. As foreign investors withdrew, banks becamemajor holders of public debt (Broner et al., 2014). Successive sovereign down-grades, ending in a sovereign debt restructuring, contributed to the collapseof the Greek banking sector. Against this background, this paper uses euroarea data to extract lessons about the processes through which sovereigns andbanks interlink. In order to do so, this paper provides a framework that relatesthe joint dynamics of �scal credit risk (Sovereign Risk) and banking credit risk(Bank Risk) to di¤erent underlying vulnerabilities and shocks. The analysis de-livers an understanding of what conditions facilitate the emergence of feedbackloops between sovereign and bank risk.
A number of recent contributions study this two-way relationship by mod-elling the common dynamics of bank and sovereign Credit Default Swaps (CDS)spreads using vector-auto regression models as in Diebold and Yilmaz (2009).According to Moody�s (2014), which studies the dynamic relation between sov-ereign and bank CDS spreads by means of a Markov switching VAR method-ology, the euro area did not su¤er one �nancial crisis, but a variety of crises,each of them with its own speci�cities. According to their results, only Irelandwitnessed a spillover of �nancial stress into sovereign stress. Instead, for Greeceand Italy their results point to the opposite feedback e¤ect. For the rest of thecountries analyzed, stress feeds back in both directions. These time series tech-niques deliver interesting indices of contagion but fall short of describing theactual channels through which such bouts of contagion take place. To bridgethis gap, this paper provides a framework conditioning the intensity of the feed-back loops on di¤erent economic factors. In doing so, similar to Acharya et al.(2013) or Mody and Sandri (2011), this paper delivers an understanding of thevulnerabilities and shocks that are fertile ground for the emergence of viciousspirals of increasing sovereign and bank risk.2
To provide estimates of how credit risk interconnectedness varies with theeconomic environment, the analysis uses detailed information on the state ofpublic �nances, the banking system and the macro economy. The paper presentsa simple econometric strategy to assess whether the sensibility of the feedbackbetween bank and sovereign risk varies with these indicators. Given the low fre-quency of macroeconomic variables and the short time series available for CDS
1 In Iceland, bank failures directly increased net public debt by 13% of GDP (Carey, 2009).2Heinz and Sun (2014) or Delatte et al. (2014) show the presence of non-linearities on
sovereign risk pricing.
1
data, the paper relies on panel data econometrics. In addition to a generalised-least-squares estimator, motivated by the high persistence of the CDS series,dynamic models are also used. The framework provides a quantitative bench-mark to measure the impact on sovereign risk of bank rescue measures, as thoseenacted by euro area governments between 2007 and 2013. Understanding thesensitivity of sovereign risk to such policies is, given the Euro Area policy set-ting, fundamental.3
The main �ndings are the following. There is a strong pass-through ofsovereign risk on bank risk. Moreover, the sovereign feedback e¤ect is quan-titatively stronger when increases in sovereign risk occur in countries with alarger stock of public debt, when the banking system exposure to the sovereignis large or when the sovereign has lost its investment grade rating. There is alsoevidence of positive spillovers from bank risk into sovereign risk. In this case,however, signi�cant pass-through appears only under speci�c macroeconomicenvironments and is signi�cantly smaller. Bank risk spillovers are signi�cantlystronger in countries where banks have bigger balance sheets and where thevolume of non-performing loans and foreign liabilities is larger. As regards therole of bank rescues, the results show that such policy operations can facilitatethe appearance of strong feedback e¤ects.
The next section summarizes the main channels through which distressspreads, as documented in the literature. The following one describes the dataand presents some preliminary evidence. The next describes the economet-ric strategy and details the main results from the analysis. The section alsopresents a detailed analysis of the e¤ect on the feedback between risks of thebank bailouts designed in Europe during the crisis. The �nal section concludes.
1 Literature review: what are the channels oftransmission?
In order to guide the analysis and help clarifying the choice of variables forcarrying out the empirical exercise, this section discusses the most relevantchannels through which �nancial and �scal stress intertwine, as identi�ed inliterature.4 These channels include the direct balance sheet interconnection, aswell as other indirect ways through which underlying vulnerabilities in eitherthe banking or public sector may materialize into twin crises.
3The European Banking Union aims to delink sovereigns and banks by allowing for bankrecapitalisation funded at the European level whenever a bank rescue risks overburdening thenational �scal position.
4Reinhart and Rogo¤ (2012) show that (i) private and public debt booms ahead of bankingcrises, (ii) banking crises, both home-grown and imported, often accompany sovereign debtcrises and, (iii) public borrowing increases sharply ahead of debt crises and (iv) it turns outthat the government has �hidden debts� (domestic public debt and contingent private debt).Closely related, Balteanu and Erce (2014) show that twin sovereign debt and banking crisesin emerging countries always combine with boom-bust patterns on the banking system.
2
A number of recent contributions study the two-way feedback between Sov-ereign and Bank stress by studying the common dynamics of bank and sovereignCDS spreads using vector-auto regression models (following the methodologyproposed by Diebold and Yilmaz (2009). While these models are extremelyuseful to understand the joint dynamics of the series, as they rely fully on thetime series dimension, they provide no economic guidance on the drivers of thefeedback e¤ects. In order to gauge an idea on the speci�c mechanisms throughwhich stress transmits, the literature has relied, instead, on pooling countrydata together. Heinz and Sun (2014) use a generalized least squares panel dataapproach to analyze sovereign CDS drivers. They show that global factors ac-count for a relevant portion of the observed variation. Acharya et al. (2013)present cross-country evidence about the potential for bank bailouts to triggera �scal crisis. Their narrative of the crisis presents three di¤erentiated peri-ods. They portray a �rst period, extending until 2007, in which sovereign riskwas never an issue within the euro area. Then, starting with the �rst bankbailouts in 2008, sovereign risk starts to surface in some parts of the MonetaryUnion as economic prospects deteriorate and public debt raises on the back ofthe support provided to a seriously deteriorated �nancial system. Since 2010,sovereign risk has become the major concern and, for some countries, implied aresurfacing of concerns regarding �nancial risk, due to the fact that a number ofbanks were either heavily exposed to the sovereign (Bruegel, 2012) or su¤eredfrom the lowering of the public guarantees provided to them (BIS, 2010). Theempirical analysis in Acharya et al. (2013) relies on the use of CDS spreads andrelates their co-movement to resolution policies and macro factors. Their resultsshow that the bailout led to an increase in sovereign risk. Moreover, they showthat, even after controlling for bank-speci�c and macroeconomic variables, thecontemporaneous relation between sovereign and bank CDS spreads remain,con�rming the existence of a sovereign bank loop. Closely related, Thukral(2013) uses a panel data framework with lagged regressors to study the role of�nancial sector variables on the determination of sovereign CDS spreads. Heconstructs a bank risk index using bank CDS spreads and �nds that the index isthe primarily statistically signi�cant determinant of sovereign risk premia evenwhen �scal variable are included, which he characterizes as bank dominance ofsovereign �nancing conditions. Mody and Sandri (2011) recognize the existenceof broadly similar sub-periods as Acharya et al. (2013), in which the feedbackbetween sovereign and bank risk changed. Instead of comparing CDS spreads,Mody and Sandri (2011) focus on sovereign spreads as a measure of the �scalrisk, and banks� stock market capitalization as a measure of risk within thebanking system. Their results, using spreads and market valuations, show thatthe euro crisis traces back to the demise of Bear Stearns. They argue that un-der the weight of increasing support for banks, sovereign spreads started to rise,especially in countries with weak growth prospects and high debt levels.
Another literature strand has delved into the role of monetary policy instrengthening the vicious relation between sovereign and bank risk. Accordingto Darraq-Pires et al. (2013), the ECB�s full-allotment liquidity policy is an
3
e¢ cient tool to stabilize spiralling feedback loops between banks and the �scalauthorities. Drechsler et al. (2013) study the reasons behind the heterogeneoustake up of long-term re�nancing operations (LTROs) among European banks.They document that banks where this take up was larger also featured largerincreases in their sovereign debt exposure.5 Drechsler et al. (2013) de�ne ahaircut subsidy associated with using government bonds as collateral with theECB, as opposed to government bonds in private repo markets. Using thissubsidy, they provide support for the hypothesis that ECB collateral policiesaction help explain the increased balance sheet interconnection between banksand sovereigns in the euro area.
As regards the main transmission channels from bank stress to the sovereign,Candelon and Palm (2010) highlight four. First, rescue plans may impair thesustainability of public �nances.6 They can include bailout money, governmentdeposits, liquidity provisioning by the central bank, public recapitalization andthe materialization of public guarantees.7 Second, if contingent liabilities mate-rialize, �scal costs are likely to be substantial. Next, the risk premium increaseseven if guarantees remain unused, raising borrowing costs for both the sovereignand the private sector (sovereign ceiling).8 Last, the downturn originated by thecredit crunch accompanying the �nancial crisis can deepen the recession, leadingto further falls in public revenues, deepening the de�cit and driving up debt.King (2009) provides an event analysis on the impact of government guaranteeson the banking system using the battery of bank rescues that took place in late2008. According to his results, the bailouts bene�ted the banks�creditors, asre�ected in falling bank CDS spreads, at the expense of equity holders, giventhat banks�stock underperformed vis-a-vis the market.
If �nancial turmoil negatively in�uences asset prices, unemployment andoutput, the direct costs increase by the impact of the crisis on tax collection andpublic expenditure. Baldacci and Gupta (2009a, 2009b) argue that sovereigndebt distress (deterioration of the �scal position) after a banking crisis is likelyto occur due to a combination of lower revenues and higher expenditures (bankrescues and outlays associated with the downturn).9 According to Honohan(2008), banking crises last 2.5 years on average, public debt increases by around30% of GDP and their estimated median �scal cost stands at 15.5% of GDP.Distress can also spread through the credit crunch created by the �nancial
5Acharya and Tuckman (2013), using data for broker-dealers in the US, show that Lender ofLast Resort activities can have the perverse side e¤ect of slowing down deleveraging, increasingilliquid leverage and the risk of default.
6Rosas (2006) studies the drivers of government intervention after banking crises. He �ndsthat authorities are more likely to bailout failing institutions in open and rich economies orif �nancial turmoil was caused by regulatory issues. On the other hand, electoral constraintsand central bank independence seem to favor bank closure.
7See Feenstra and Taylor (2008) or Reinhart and Rogo¤ (2011).8Laeven and Valencia (2011) show that blanket guarantees increase the �scal costs of
banking crises, but this can also be because they are set in place during severe crises.9Baldacci and Gupta (2009) argue that �scal expansions do not improve the growth outlook
by themselves and lead to higher interest rates on long-term government debt. They identify atrade-o¤ between boosting aggregate demand (short-run) and productivity growth (long run).
4
crisis. As credit falls or becomes more expensive, the economy is likely tosu¤er a drop in GDP growth. This might put additional pressure on the �scalposition through its impact on tax revenues, likely to be lower as activity falls.10
Relatedly, Laeven and Valencia (2011) focus on the impact of �nancial sectorinterventions on the capacity of the �nancial system to provide credit. Theirresults show that �rms dependent on external �nancing bene�ted signi�cantlyfrom bank recapitalization operations. However, as documented in Acharya, ifthe sovereign becomes overburdened, the value of the public guarantees falls,deepening the interconnection of stress. Kollmann et al. (2012) also focuson the impact of bank rescues. Their message is positive and highlights theability of bank rescue operations to improve macroeconomic performance. Still,while they show that bank rescues raise investment, in line with the evidence inBroner et al. (2014) or Popov and Van Horen (2013), they �nd that sovereigndebt purchases by domestic banks lead to a crowding out of private investment.Gray and Jobst (2011) and Gray et al. (2013) present a less benign exerciseshowing the potentially high impact on �scal risk associated to the existence ofcontingent liabilities.
Finally, if uncertainty augments the crisis could lead to a sudden stop ofcapital in�ows. In this line, Reinhart and Rogo¤ (2008) argue that bankingcrises often follow credit booms and high capital in�ows. Moreover, they �ndthat periods of high international capital mobility gave rise to banking crisesin the past. Cavallo and Izquierdo (2009) provide further evidence showingthat, after �nancial crises in emerging markets, capital �ows may collapse formonths or years potentially triggering a solvency crisis. Indeed, as argued byObstfeld (2011) when discussing the role of international liquidity in the recentdebt crisis, �. . . gross liabilities, especially those short-term, are what matter�.Van Rixtel and Gasperini (2013) show that sovereign risk, as measured by thesovereign swap spreads, has shown in some periods a strong correlation with thethree-month USD Libor-OIS, a sign that borrowing strains in foreign currencyfor banks a¤ect the creditworthiness of the sovereigns.
In turn, a number of transmission channels of a �scal crisis on the broadereconomy can be traced through the domestic �nancial system.11 Wheneverassets need to be written o¤ or rescheduled, domestic banks are usually the�rst in line to take a hit. Along these lines, Noyer (2010), argues that banks�holdings of defaulted government bonds might lead to large capital losses andthreaten the solvency of elements of the banking sector. IMF (2002) provides acomprehensive overview of the e¤ects of four sovereign restructurings (Ecuador,Pakistan, Russia and Ukraine) on the domestic banking sector. The paper doc-uments the extent of direct losses from banks�holdings of government securities,an increase in the interest rates on liabilities not matched by increased returnson assets (on the contrary, in this context government securities usually o¤ernon-market rates), and an increase in the rate of non-performing loans increases,as higher �nancing costs lead to corporate bankruptcies. Similarly, Erce (2012)
10See De Paoli et al. (2009) or Feenstra and Taylor (2008).11See IMF (2002) or Reinhart and Rogo¤ (2012).
5
suggests that the degree of bank intermediation and the banking system expo-sure to the sovereign strongly in�uence a debt crisis ripple e¤ect on the realeconomy. In addition, authorities often react to debt problems by coercing do-mestic creditors to hold government bonds in non-market terms (Diaz-Cassouet al., 2008).12 While this keeps borrowing costs low, a government default maytrigger a banking crisis.13 In Darraq-Pires et al. (2013) the positive connec-tion between sovereign and bank risk is due to banks investing in governmentsecurities to hedge future liquidity shocks. Along these lines, Angeloni andWol¤ (2012) assess the impact of sovereign bond holdings on the performanceof banks during the euro area crisis using individual bank data and sovereignbond holdings. They �nd that peripheral sovereign bonds a¤ect banks�stockmarket valuations heterogeneously. While Italian, Irish and Greek debt appearto have negatively a¤ected the market valuation of the banks holding them,such an e¤ect is not signi�cant for other peripheral sovereign debt, most no-tably, Spanish.14 Acharya et al. (2013), document the high exposure of theirsample banks to their own sovereign, which according to their theory should bea main channel through which stress feeds back.15
Beyond this direct balance sheet e¤ect, the ensuing �scal contraction maylead to reduced activity, a¤ecting banks�pro�ts and further damaging the �nan-cial system. Moreover, a credit crunch may worsen the economic downturn, asbanks reduce lending due to capital losses and to the increase in uncertainty thatcomes with a sovereign debt default (Panizza and Borenzstein, 2008). Popovand Van Horen (2013) focus on the feedback from sovereign risk into bankingrisk by assessing the extent to which increasing holdings of distressed sovereignbonds limit the banks�ability to extend loans to the private sector, furtheringthe vicious feedback loop by limiting the growth potential of the economy. Theydocument a stronger reallocation away from domestic lending in the periphery.A similar crowding out e¤ect is present in Broner et al. (2014), who present abattery of stylized facts for the euro area, including both an increase in sovereignbond holdings by banks and a simultaneous drop in �nancing to the private sec-tor.16 Corporate borrowers and banks may face a sudden stop after a sovereigndefault even if their exposure to government bonds is limited. Gennaioli et al.(2010) and Erce (2012) argue that sovereign defaults trigger capital out�owsand credit crunches. An additional pressure to curtail lending might come from
12Das et al. (2012) argue that regulatory factors could lead to further balance sheet in-tertwining. In Livshits and Schoors (2009), as public debt becomes risky, governments haveincentives to not adjust prudential regulation.13 In past crises, prudential regulation treated government bonds as risk-free despite default
expectations were not zero (IMF, 2002). According to Castro and Mencia (2015), a similarphenomenon has been at play in the Eurozone14A caveat of this analysis is that data stops in mid-2012, before the height of stress in Italy
and Spain.15Among other things, the paper assesses the extent to which reduced sovereign ratings
a¤ected the banks CDS through their e¤ect on the public guarantees.16These papers present a nuanced view of domestic purchases of public debt. Others have
found positive e¤ects. According to Asonuma et al. (2015) and Andritzky (2012), domesticbank purchases of sovereign bonds help stabilize sovereign funding costs.
6
the fact that the economic uncertainty may lead to deposit runs or a collapseof the inter-bank market (Panizza and Borenzstein, 2008). Finally, sovereignrating downgrades further limit banks�access to foreign �nancing, leading tosudden stops or higher borrowing costs (Reinhart and Rogo¤, 2012).
2 Data
On the sovereign front, some authors have measured credit risk using creditratings (Correa et al., 2012) or bond spreads (Mody and Sandry, 2011). In turn,bank risk proxies previously used include credit ratings (Correa et al., 2012) andthe stock market behavior (Angeloni and Wol¤, 2012). The analysis here followsa recent strand of the literature that has opted for using credit default swaps(CDS). By design, CDS contracts shield the holders from events of default, soare the �nancial instruments most related to credit risk. Importantly, althoughthe data spans back a little less than a decade, CDS markets are relativelyliquid.17 Monthly data for 5-year CDS contracts for both individual banksand sovereigns comes from Bloomberg and DataStream. For sovereign CDSdata, in most countries the information spans back to late 2005. In order to beable to assess the various twists observed during the crisis, countries for whichsovereign CDS data was missing prior to 2008 (Cyprus and Luxembourg) wereexcluded from the sample. In turn, the above-cited sources returned activeCDS contracts for 48 banks in the euro area. Unfortunately, prior to 2007,the coverage was less homogeneous. When considering together the coverage ofboth banks and sovereign entities, su¢ ciently large series were available for 10euro area countries: Germany, Italy, France, Spain, Ireland, Greece, Portugal,Belgium, Netherlands and Austria.18
As in Acharya et al. (2013), to have a system-wide measure of bank stress,individual bank CDS data is aggregated in a country-speci�c bank risk index.De�ning the CDS of bank j 2 J from country i at time t by Bank CDSjit andthe corresponding weight as wjit,country�s i Bank Risk Index is
BankRiskit =X8j2J
wjitBankCDSjit
From the various weighting schemes available, for simplicity, this paper useswjit =
1J :19
The econometric exercise controls for various macroeconomic, �nancial andglobal factors. Data on sovereign ratings comes from Fitch. Data on the banks�balance sheets come from Haver Analytics, the European Central Bank, the
17An important limitation of CDS data relates to the existence of counterparty risk. Thelack of detailed data on CDS counterparties prevents from controlling for this potential bias.18There is no CDS data for Finnish banks, preventing its inclusion in the analysis.19Banks weights could be set according to their market capitalization or total assets. While
the �rst option above focuses on private capital, depending on the extent of bank nationali-sation, the second can be more adequate.
7
Bank for International Settlements and the IMF�s Financial Stability Indica-tors.20 The series included are: total assets, exposure to the general government,funding from the central bank, foreign assets and liabilities, non-performingloans, return on assets and equity ratio. Macroeconomic data (unemployment,in�ation, nominal GDP growth, �scal de�cit, current account and public debt)was obtained from Haver Analytics.21 The Itraxx �nancial Junior and VIXindex come from Bloomberg.
3 Preliminary Evidence
Figures 1 and 2 (in the Appendix) provide a bird�s eye view on the behavior ofthe risk series. Figure 1 portrays the behavior of sovereign and bank risk from anaggregate perspective. Euro area wide sovereign stress is proxied using a simpleaverage of sample countries�sovereign CDS. The Itraxx Junior represents bankrisk. In turn, Figure 2, shows the behavior of sovereign and bank on a country-by country basis.
As a reminder of the importance of policy action, the shadowed areas inFigure 1 represents two periods of marked policy activism. The �rst depictsthe two months of 2008 in which most sample countries enacted programs ofsupport for their �nancial systems. Remarkably, even at the low frequencyemployed here, the very speci�c dynamics ongoing during the third quarter of2008 are still apparent. On the back of the public guarantees, the bank creditrisk decreased markedly. However, simultaneously, the sovereign CDS startedto pick up. According to Acharya et al. (2013), the increasing sovereign CDSre�ected market fears regarding the just absorbed liabilities. The second periodshadowed in Figure 1 corresponds to that following the ECB announcement ofthe Outright Monetary Transactions (OMT) instrument (August 2012). Whileit is not apparent that such policy action changed the correlation, Figure 1shows a change in risk dynamics. Since then, both risk indicators have trendeddown. Another way to look at time patterns for the correlation between the riskvariables comes from comparing sub-periods. This is done in Table 1 below.
20 IMF�s FSI indicators (non-performing loans, return on assets and equity ratio) are avail-able only since 2008.21Converse to the literature on sovereign spreads that focuses on real GDP, nominal GDP
is used given its relevance in markets�assessment of debt sustainability. The debt and �scaldata refers to the General Government. These variables, as GDP, are available only on aquarterly basis. They have been linearly interpolated into monthly frequency.
8
In periods 2 and 3 (bail-out and �scal activism), the correlation observedpreviously broke down. Remarkably, since the inception of the OMT, the corre-lation is back to its pre-crisis value.22 Following, Broner et al. (2014) narrativeof the crisis, further insights into the dynamic relation of the risk indicators canbe gained by breaking the euro area into a core and a periphery. This is doneby running the following regressions
Risk_Ait = �i +X
p2(1;5)
�p � Period p dummy �Risk_Zit�1 + "it; (1)
Risk_Ait = �i +X
r2(core;periph:)
�r � region r dummy �Risk_Zit�1 + �it(2)
where Risk_Ait and Risk_Zit stand, interchangeably, for country�s i sov-ereign and bank risk. Within regression (1) the feedback e¤ect from one riskto the other is allowed to depend on the speci�c periods described in Table1. In turn, within regression (2) the coe¢ cients are allowed to di¤er di¤er be-tween core and peripheral countries. The results are presented in Table 2 in theAppendix. The European crisis period (January 2010-August 2012) featureda particularly large degree of pass-through from bank risk into sovereign risk.Notably, feedback loops are not too di¤erent in peripheral and core countries.If anything, bank risk has a stronger pass-through e¤ect on sovereign risk in pe-ripheral economies. Overall, there is some evidence of the correlation betweenrisk indicators having diverged across time and regions. The rest of the paperattempts to connect this time and spatial variation in risk to the dynamics ofthe underlying macroeconomic conditions.
4 Econometric Analysis
This section presents a panel data model of the feedback loop for each riskvariable.23 As in Thukral (2013) or Heinz and Sun (2014), the starting point is
22To complement the data description, Table A1 in the Appendix presents summary statis-tics for the full sample and for the core and periphery subsamples.23The low number of observations calls for pooling country data to take advantage of both
time series and cross-country variation and for keeping the model as parsimonious as possible.
9
a Generalized Least Squares (GLS) estimator, using the CDS variables in levels.Following the literature, in addition to the risk indicators, the model controlsfor �nancial, global, macroeconomic, and contagion e¤ects:
Risk_Ait = �Ai + �ZARisk_Zit�1 +�AAXA
it�1 +�ZAXZit�1 +�GAX
Git�1 + "
Ait
Within this framework, the coe¢ cient �ZA measures the extent to whichRisk Z feeds into Risk A. In addition, the model controls for the primarydeterminants of Risk A (XA
it ) and Risk Z (XZit ). When dealing with the sov-
ereign risk model, XAit collects the macro variables and X
Zit collects the banking
sector variables. When dealing with the bank risk model, this reverses. Thevariable �Aicollects country-speci�c characteristics. Euro area sovereign debtmarkets have been subject to recurrent bouts of dramatic co-movement duringthe crisis, which a number of commentators have associated with contagion.24
This cross-sectional correlation can bias the standard errors, making the esti-mations less reliable. To address this issue the model controls for global andcontagion factors (XG
it ). To gauge the relative importance role of the di¤erentsets of covariates, they are included and discussed in steps.
Additionally, the high degree of persistence of the CDS series raises con-cerns about the robustness of the results. To address this concern the modelincorporates dynamic e¤ects by including a lag of the dependent variable,
(1� AL)Risk_Ait = �Ai + �ZARisk_Zit�1 + �Xit�1 + "Aitwhere L is the lag operator, A is the autoregressive coe¢ cient of Risk A,
� = [�AA;�ZA;�GA] and Xit�1 = [XAit�1; X
Zit�1; X
Git�1]. The bias (Nickel bias)
introduced by the dynamic element is tackled by using system-GMM (Arellanoand Bover, 1995), which relies on the use of internal instruments (lagged levelsand di¤erences of the endogenous and predetermined variables).
4.1 Sovereign Risk Model
In a �rst step, similar to D�Agostino and Ehrmann (2014), the model only usesthe macro factors. The variables included are: debt to GDP, �scal balance,�nancial account, GDP growth, unemployment and in�ation.25 The results(Column 1, Table 3) are broadly in line with previous literature. Remarkably,the �scal balance shows no signi�cant relation with sovereign risk. Next, toassess the importance of banking factors for the pricing of sovereign risk, themodel also includes the bank risk determinants. Following the literature, the re-gressors include: loan quality (non-performing loans to total loans), pro�tability
Signi�cant gaps in Greek data preclude its use on the econometric part.24According to Alter and Beyer (2013) or Broto and Perez-Quiros (2013) contagion played a
non-negligible role in peripheral countries. Heinz and Sun (2014) �nd that shocks to Spanishand Italian CDS delivered the largest spillovers.25The Breusch-Pagan Lagrange Multiplier test strongly supported the inclusion of random
e¤ects.
10
(return on assets), bank capital (tangible common equity ratio), the home biasin the banks�portfolio (domestic assets as a % of total assets), the exposure topublic entities (private assets over total assets) and a measure of funding stabil-ity (assets to deposits). The results, in column 2, serve as test for the �nancialdominance hypothesis (Thuckar, 2013). While banking variables heavily in�u-ence the behavior of sovereign risk, converse to Thuckar (2013), macroeconomicfactors play a dominant role.26
The next step adds BankRiskit to the framework. The coe¢ cient associatedwith the bank risk indicator measures the feedback from bank into sovereign risk.Column 3 presents the results for this model. There is a positive and signi�cantrelation between bank and sovereign risk. For every 10 basis points (bps) in-crease in bank risk, sovereign risk increases by 4.2 bps in the following month.This is a large degree of pass-through. To lower the degree of commonality in theerror terms, the model also controls for global shocks and potential contagione¤ects.27 To proxy contagion, the model includes the average of the sovereignCDS for other euro area countries. In turn, the model includes the VIX index toproxy for global shocks. Column 4 from Table 3 presents the results. While theVIX Index does not appear to have a signi�cant relation to sovereign risk, thecontagion indicator presents a highly signi�cant positive relation with sovereignrisk. Controlling for global and contagion e¤ects does not alter the signi�canceof pass-through, although the size of the coe¢ cient becomes smaller (3.1 bpsincrease in sovereign risk for every 10 bps increase on bank risk).28
Finally, column 5 presents a dynamic version of the sovereign risk model.As detailed above, the model is estimated using system GMM.29 The dynamicelement is large (close to unity) and highly signi�cant. Remarkably, while thepass-through from bank to sovereign risk remains signi�cant, the sign reverses.According to the results, for every 10 bps increase in bank risk, sovereign riskdecreases by 0.9 bps.
4.2 Bank Risk Model
Following similar steps, the bank-related variables are included �rst. Next, themacroeconomic controls are introduced. Global shocks are again proxied withthe VIX. Instead, contagion e¤ects are now accounted for using the Itraxx Juniorindex. Finally, the dynamic version of the model, including the lagged value ofbank risk, is estimated. Table 4 presents the results for these models.
As shown in Columns 1 and 2 of Table 4, banks with a larger home biasand larger private sector credit face larger bank risk. Non-performing loans areassociated, as expected, with higher bank risk. Interestingly, a lower ratio of
26The regression�s R-squared increases by more than 50% after adding the bank variables,but still gives macro factors a larger weight in explaining the sovereign risk variance.27A Pesaran test on the model´ s residuals shows a signi�cant degree of spatial correlation.28The results (available under request) using a two-step Driscoll-Kraay correction for cross-
sectional correlation are almost undistinguishable.29Both the Sargan endogeneity tests and the Di¤erence-in-Hansen tests of exogeneity tests
validate the instruments.
11
assets to deposits and higher bank capital are associated with larger levels ofstress. This result could be re�ecting the fact that banks located in countrieswith stronger sovereigns have less need to build their own capital cushions (asin De Grauwe and Ji, 2013).30 Column 3 shows the results for the modelincluding the lagged value of sovereign risk. The feedback coe¢ cient is, again,highly signi�cant (0.53). In turn, as expected, larger values for the Itraxxand VIX Indices associate with more bank risk (column 4). Contagion acrossbanks is a signi�cant phenomenon. Finally, column 5 of Table 4 presents theestimates for the dynamic model of bank risk. The coe¢ cient of main interest,the one associated with the sovereign risk indicator, is positive and signi�cant.According to the results, a 10 bps increase in sovereign risk leads to a 0.8 bpsincrease in bank risk.
4.3 A cheat impulse-response
Combining the pass-through coe¢ cients obtained from the sovereign and bankrisk models, one can recoup the dynamic response of sovereign and bank riskto shocks to one another. The �gures below present a graphical representationof shocking such system of equations with a 50 bps shock to sovereign risk (leftchart) and to bank risk (right chart).
Figures 3.1 and 3.2 illustrate the di¤erent form that average feedback e¤ectstake. On the one hand, there is a strong positive feedback arising from sovereignshocks (Figure 3.1). On the other, there is no evidence of a feedback loop frombank risk into sovereign risk. Quite the opposite, bank risk shocks induce amilder and negative reaction of sovereign risk (Figure 3.2).
30 In unreported estimates using the Driscoll-Kraay correction, the results are qualitativelyidentical
12
5 Digging into the Sources of Feedback Loops
The relation between both risks might depend on the underlying economic and�nancial environment. For instance, according to Acharya et al. (2013) or Mar-tin et al. (2014), explicit and implicit balance sheet interrelations can powerfullyamplify feedback loops. This section tests what conditions a¤ect the intensityof the pass through by incorporating interactions between the risk measure andother variables,
where Fit�1 is the factor interacting with the Risk Z. Within this frame-work, the feedback between risks becomes:
@Risk_Ait@Risk_Zit�1
= �ZA + �FZAFit�1
The sovereign risk model with interactions is estimated for the followingvariables: size of the banking system (Gennaioli al., 2014), banks�foreign liabil-ities (Cavallo and Izquierdo, 2009) and banks�non-performing loans (Acharyaet al., 2013).31 In turn, the candidate variables for a¤ecting the feedback fromthe sovereign to the banks are public debt to GDP (Mody and Sandry, 2011),banks�balance sheet exposure to the sovereign (Angeloni and Wol¤, 2012), andthe investment grade status of sovereign debt (Correa et al., 2012). Table 5(sovereign risk) and Table 6 (bank risk) contain the result.
Table 5 vindicates the validity of most of the above-mentioned channelsof transmission. It shows that the three interactions present signi�cant posi-tive spillovers from bank to sovereign risk. The pass-through of risk becomesstronger where the volume of non-performing loans and banks�foreign liabili-ties are larger. Conversely, there is no evidence that, where banks have biggerbalance sheets, the feedback e¤ect is stronger.
In turn, Table 6 shows that the feedback from sovereign into bank risk isstronger the larger the stock of public debt and larger banking system exposureto the sovereign. The results also show a signi�cantly stronger pass-through ofsovereign risk when the sovereign rating is below investment grade.32 When asovereign rating falls outside the investment grade category, it loses a large poolof potential investors, a¤ecting negatively sovereign risk.
5.1 Economic signi�cance
To grasp the economic relevance of these results, Figures 4.1 and 4.2 depictvarious e¤ects in basis points (bps). Figure 4.1 shows how the pass-through onto
31All the variables are measured as percentage of GDP to make them relative to the au-thorities�potential.32This is despite the fact that the adjustments to the ECB�s collateral policy during the
crisis (Eberl and Webber, 2014) ameliorated the impact of not having an investment grade.
13
sovereign risk of a 100 bps increase in bank risk depends on di¤erent values ofFit. Figure 4.2 does the same for the e¤ect on bank risk of a 100 bps increase insovereign risk. The �gures compare the e¤ects at the minimum and maximumvalues within sample of the corresponding indicators.
Some of the conditional risk dynamics are economically very sizeable. Forinstance, Figure 4.1 shows that a 100 bps increase in bank risk does not leadto a positive feedback on sovereign risk even if the banking system size is atits maximum within the sample. The feedback is, instead, very large when theasset quality of the banks, as measured by the share of non-performing loans(NPLs), is high. While for the lowest level of NPLs there is no positive feedbacke¤ect, at the maximum value within sample, the e¤ect is well above 150 bps.
Similarly, when banks�foreign liabilities are large, there is a sizeable positivefeedback e¤ect of bank risk to sovereign risk. In turn, Figure 4.2 shows therelevance of the balance sheet exposure to the sovereign in the transmissionof stress. Faced with an increase in sovereign risk of 100 bps, banking systemsholding the lowest level of exposure face an 18 bps increase in their risk. Instead,banks with larger exposures face an increase of 80 bps. The feedback e¤ect canalso grow considerably in the presence of large public debt stock (up to 62 bps),and when the sovereign has lost its investment grade (40 bps).
6 Bank Rescues and the Feedback Loop
This section uses the sovereign risk model to assess quantitatively the e¤ectthat bank rescue operations can have on the feedback from bank into sovereignrisk. According to Acharya et al. (2013), the rescue packages enacted by euroarea governments to �ght o¤ the �nancial crisis generated a risk transfer. Assovereigns began to support their banks, investors became more con�dent aboutbanks. This led to a lowering of banks�CDS spreads. Unfortunately, in somecases, the weight governments had to lift pushed up sovereign risk, facilitating
14
the emergence of a perverse feedback loop.33 To limit extreme forms of this risktransfer, the euro area authorities devised a tool to assist banks directly usingthe European Stability Mechanism (ESM, 2014).34 Implementing this policyrequires determining when a sovereign might not be able to do it on its own.The analysis focuses on direct exposures and contingent liabilities.35
Figure 5 provides a dynamic representation of the e¤ects of a shock to bankrisk when the sovereign has bailed out the banks using an amount equal to theaverage �scal cost of bank crises (15% of GDP) in Laeven and Valencia (2011).
In line with Acharya et al. (2013) risk-transfer hypothesis, the results, pre-sented in Table 7, point to a signi�cantly larger pass-through of bank risk intothe sovereign for those economies where the authorities more heavily supportedtheir banking system. According to the results, given a size of the bailout equalto 15% of GDP, for every 100 bps increase in bank risk, sovereign risk increasesby 11 bps within a year. As shown in columns 3 and 4, this e¤ect becomes moresizeable for countries where the banks have a larger amount of foreign liabilitiesor a larger balance sheet exposure to the sovereign.
7 Conclusions and Policy implications
This paper has analyzed the factors associated with the emergence of perversespirals of sovereign and bank stress. Using a dynamic panel data model, ituncovers underlying vulnerabilities that reinforce the process where shocks to
33Alter and Beyer (2013) �nd that, in Spain, the nationalization of Bankia led to an increaseon spillovers.34Direct recapitalisation is provided if a sovereign cannot provide support without triggering
a �scal crisis.35The data, in an annual format, comes from the European Commission.
15
a country´s �scal health contaminate the �nancial sector. Countries wherepublic debt is larger, and where domestic banks have a larger exposure to theirown sovereign, face stronger feedback loops from sovereign into bank risk. Thesame goes for countries losing their investment grade status. On the other, theanalysis also identi�ed factors associated with an elevated transmission of bankdistress to the sovereign. In countries where banks are larger, funded with moreforeign credit and face more non-performing loans, the feedback from bank riskinto sovereign risk is stronger.From an economic policy perspective, these results can help in monitoring
the build-up of �scal weaknesses and the robustness of the �nancial system to�scal shocks. Additionally, the new framework to handle banking crises in theeuro area implies that, if the foreseen bail-in of the bank�s private creditors isnot enough, individual banks could be rescued directly by the o¢ cial sector.For such direct recapitalization to happen, it has to be the case that the coun-try could endanger its sustainability if supporting the bank alone. This paperinforms this process by studying the circumstances in which �nancial rescuesmight overburden the sovereign.
References
[1] Acharya, V., I. Dreschlet, and P. Schnabl (2013), A Pyrrhic Victory: BankBailouts and Sovereign Credit Risk, Mimeo.
[2] Acharya, V., and B. Tuckman (2013), Unintended Consequences of LOLRFacilities: The Case of Illiquid Leverage, NBER Working Paper 19773.
[3] D�Agostino, A., and M. Ehrmann (2013), The pricing of G7 Sovereign bondspreads: The times, they are a-changing, Journal of Banking and Finance,forthcoming.
[4] Alessandri, P. and A. Haldane (2009), Banking on the State, Bank of Eng-land, mimeo.
[5] Alter, A. and A. Beyer (2013), The Dynamics of Spillover E¤ects duringthe European Sovereign Debt Turmoil, Mimeo.
[6] Andritzky, J. (2012), Government Bonds and Their Investors: What arethe Facts and Do They Matter?, IMF Working Paper 12/158.
[7] Angeloni, C., and G. Wol¤, 2012, Are Banks A¤ected by Their Holdingsof Government Debt?, Bruegel Working Paper 2012/07.
[8] Arce, O., Mayordomo, S., and J. Pena (2012), Credit-risk valuation inthe sovereign CDS and bond markets: Evidence from the euro area crisis,CNMV Working Paper Series, No. 53.
[9] Arellano, C. and N. Kocherlakota (2008), Internal Debt Crises and Sov-ereign Defaults, NBER Working Paper 13794.
16
[10] Asonuma, T., Bakhache, S., and H. Hesse (2015), Is Banks�Home BiasGood or Bad for Public Debt Sustainability?, IMF Working Paper 15/44.
[11] Baldacci, E. and S. Gupta (2009a), Fiscal Expansions: What Works, Fi-nance & Development, Volume 46, Number 4.
[12] Baldacci, E., C. Mulas-Granados and S. Gupta (2009b), How E¤ective isFiscal Policy Response in Systemic Banking Crises?, IMF Working Papers09/160.
[13] Balteanu, I., and A. Erce, 2014, Bank Crises and Sovereign Defaults inEmerging Markets: Exploring the Links. Bank of Spain, Working Paper1414.
[14] Broner, F., Erce, A., Martin, A., and J. Ventura (2014), Sovereign DebtMarkets in Turbulent Times: Creditor Discrimination and Crowding Out,Journal of Monetary Economics, Volume 61.
[15] Broto, C. and G. Perez Quiros (2013), Disentangling contagion among Sov-ereign CDS spreads during the European debt crisis, Bank of SpainWorkingPaper Series No. 1314.
[16] Candelon, B. and F.C. Palm (2010), Banking and Debt Crises in Europe,The Dangerous Liaisons?, CESifo Working Paper No. 3001.
[17] Caprio, G. Jr. and P. Honohan (2008), Banking Crises, Institute for Inter-national Integration Studies Discussion Paper Series no. 242.
[18] Castro, C., and J. Mencia (2014), Sovereign Risk and Financial Stability,Revista de Estabilidad Financiera No. 26, Bank of Spain.
[19] Cavallo, E. and A. Izquierdo (2009), Dealing with an International CreditCrunch: Policy Responses to Sudden Stops in Latin America, Inter-American Development Bank, mimeo.
[20] Carey, D. (2009), Iceland: The Financial and Economic Crisis, OECDEconomics Department Working Papers, No. 725, OECD Publishing.
[21] Correa, R., Kuan-Hui, L., Sapriza, H., and G. Suarez (2012), SovereignCredit Risk, Banks�Government Support, and Bank Stock Returns Aroundthe World, Federal Reserve Board, International Finance Discussion papersNo. 2012-1069
[22] Das, U.S., Oliva, A.O., and T. Tsuda (2012), Sovereign Risk: A Macro-Financial Perspective, Public Policy Review Vo. 8, No. 3 (Ministry of Fi-nance, Japan).
[23] Delatte, A. L., Fouquau, J. and R. Portes (2014), Nonlinearities in Sov-ereign Risk Pricing: The Role of CDS Index Contracts. OFCE WorkingPaper 2014-08.
17
[24] De Grauwe, P., and Y. Ji (2013), Strong Governments, Weak Banks, CEPSPolicy Brief No. 305.
[25] De Paoli, B., Hoggarth, G. and V. Saporta (2009), Output costs of sovereigncrises: some empirical estimates, Bank of England working papers no. 362.
[26] Diaz-Cassou, J., A. Erce and J. Vazquez (2008), Recent episodes of sov-ereign debt restructuring: A case-study approach, Bank of Spain, Occa-sional Document 0804.
[27] Diebold, F., and K. Yilmaz (2009), Measuring �nancial asset return andvolatility spill overs, with an application to global equity markets, EconomicJournal, Issue 119.
[28] Dreschler, I., Drechsel, T., Marques-Ibanez, D., and P. Schnabl (2013),Who Borrow from the Lender of Last Resort?, Mimeo.
[29] Eberl, J., and C. Webber (2014) ECB Collateral Criteria: A NarrativeDatabase 2001-2013, Ifo Working Paper No. 174
[30] Erce, A. (2012), Selective Sovereign Defaults, Federal Reserve Bank of Dal-las, Globalization and Monetary Policy Institute, Working Paper No. 127.
[31] European Commission (2009), Public Finance Report in EMU-2009, Euro-pean Economy No. 5.
[32] European Stability Mechanism (2014), ESM direct bank recapitalisationinstrument adopted, Press release no 18/2014
[33] Feenstra, R.C and A. Taylor (2008), International Economics, Worth Pub-lishers.
[34] Gennaioli, N., Martin, A. and S. Rossi (2014), Sovereign Default, DomesticBanks and Financial Institution, Journal of Finance, forthcoming.
[35] Gray, D. and A. Jobst (2011), Modelling systemic �nancial sector and sov-ereign risk, Sveriges Riskbank Economic Review, 2011:2.
[36] Gray, D., Gross, M., Paredes, J., and M. Sydow (2013), Modelling Banking,Sovereign and Macro Risk in a CCA Global VAR, IMF Working Paper13/218.
[37] Heinz, F. and Y. Sun (2014), Sovereign CDS Spreads in Europe: TheRole of Global Risk Aversion, Economic Fundamentals, Liquidity, and Spillovers, IMF Working Paper 14/17.
[38] Honohan, P. (2008), Risk Management and the Costs of the Banking Crisis,The Institute for International Integration Studies Discussion Paper Seriesno. 262.
18
[39] IMF (2002), Sovereign Debt Restructurings and the Domestic EconomyExperience in Four Recent Cases, Policy Development and Review Depart-ment.
[40] King, M. (2009), Time to buy or just buying time? The market reactionto bank rescue packages, BIS Working Paper No. 288.
[41] Kollmann, R., Roeger, W., and J. in�t Veld (2012), Fiscal Policy in a Fi-nancial Crisis: Standard Policy vs. Bank Rescue Measures, Ecares WorkingPaper 2012-006.
[42] Laeven, L. and F. Valencia (2011), The Real E¤ects of Financial SectorInterventions During Crises, IMF Working Paper 11/45.
[43] Livshits, I. and K. Schoors (2009), Sovereign Default and Banking, mimeo.
[44] Mody, A., and D. Sandri, 2011, The Eurozone Crisis: How Banks Came tobe Joined at the Hip, IMF Working Paper 11/269.
[45] Moody�s (2014), European Sovereign Debt and Banking Crises: Contagion,Spill overs and Causality, Moody�s Investors Service, Credit policy.
[46] Noyer, C (2010), Sovereign crisis, risk contagion and the response of thecentral bank, mimeo.
[47] Panizza, U and E. Borensztein (2008), The Costs of Sovereign Default, IMFWorking Paper 08/238.
[48] Reinhart, C.M. and K.S. Rogo¤ (2008), Banking Crises: An Equal Oppor-tunity Menace, NBER Working Papers 14587.
[49] Reinhart, C.M. and K.S. Rogo¤ (2012), From Financial Crash to DebtCrisis, American Economic Review, 101(5).
[50] Rosas, G. (2006), Bagehot or Bailout? An Analysis of Government Re-sponses to Banking Crises, American Journal of Political Science Vol. 50No.1, pages 175-191.
[51] Popov, A., and N. Van Horen (2013), The impact of sovereign debt exposureon bank lending: Evidence from the European debt crisis, DNB WorkingPaper No. 382.
[52] Thukral, M. (2013), Bank dominance: Financial sector determinants ofsovereign risk premia, Mimeo.
[53] Van Rixtel, A., and G. Gasperini (2013), Financial Crises and bank funding:recent experience in the euro area, BIS Working Papers No. 406.
19
Appendix
Figure 1: Sovereign and Bank Risk in the Euro Area
Data Series Source Frequency
Local currency rating Fitch Monthly
Harmonized CPI Index Haver Analytics Monthly
Nominal GDP Haver Analytics Quarterly
Financial Account Balance Haver Analytics Quarterly
Harmonized Unemployment Rate Haver Analytics Monthly
General Government
Nonconsolidated DebtHaver Analytics Quaterly
General Government: Net
Lending/BorrowingHaver analytics Quarterly
Banking System Balance Sheet Haver Analytics Monthly
VIX index CBOE Monthly
Itraxx Junior Financial Indices Bloomberg Monthly
Central Bank Lending Individual Central Banks Monthly
Bank rescue operations (l iabilities
and contingent l iabilities)European Comission Annual
Sovereign 5-year CDS spreads Bloomberg and Datastream Monthly
Bank 5-year CDS spreads Bloomberg and Datastream Monthly
Variables included in the analysis: Main features
Figure 2: A bird’s eye view of Sovereign and Bank risk
0
100
200
300
400
0
100
200
300
400
2005m1 2010m1 2015m1
2005m1 2010m1 2015m1 2005m1 2010m1 2015m1
Austria Belgium France
Germany Netherlands
5-YEAR CDS SOVEREIGN 5-YEAR CDS BANK INDEX
0
50
010
00
15
00
20
00
0
50
010
00
15
00
20
00
2005m9 2014m1 2005m9 2014m1
Ireland Italy
Portugal Spain
5-YEAR CDS SOVEREIGN 5-YEAR CDS BANK INDEX
Variable Observations Mean Std. Dev. Min Max Observations Mean Std. Dev. Min Max Observations Mean Std. Dev. Min Max
Data runs from September 2007 unti l January 2014. Core countries are Germany, France, Belgium, Austria and Netherlands . Periphery countries include Ireland, Ita ly, Portugal , Greece and Spain.
Table A1. Summary statistics by geographical area: Core versus periphery
Dep. Var: Sovereign Risk Ful l Sample Core vs Periphery
Bank Risk Index (during Period 1) 8.74E-02
[0.09]
Bank Risk Index (during Period 2) 2.54e-01**
[0.11]
Bank Risk Index (during Period 3) 2.52e-01***
[0.03]
Bank Risk Index (during Period 4) 6.04e-01***
[0.02]
Bank Risk Index (during Period 5) 3.66e-01***
[0.02]
Bank Risk Index (i f core country) 4.80e-01***
[0.06]
Bank Risk Index (i f periphera l country) 5.49e-01***
[0.02]
Constant 2.77e+01** 6.64
[11.48] [18.48]0 0
Observations 890 890
R-squared 0.57 0.47
Ful l Sample Core vs Periphery
Dep. Var: Bank Risk
Sovereign Risk (during Period 1) 2.25e+00***
[0.85]
Sovereign Risk (during Period 2) 2.72e+00***
[0.74]
Sovereign Risk (during Period 3) 2.26e+00***
[0.14]
Sovereign Risk (during Period 4) 1.04e+00***
[0.03]
Sovereign Risk (during Period 5) 1.19e+00***
[0.06]
Sovereign Risk (i f core country) 1.16e+00***
[0.11]
Sovereign Risk (i f periphera l country) 1.01e+00***
[0.03]
Constant 7.46e+01*** 9.93e+01
[13.09] [61.77]
Observations 887 887
R-squared 0.53 0.49
Table 2. Bank and Sovereign risk loops by periods and regions
Standard errors in brackets . *** p<0.01, ** p<0.05, * p<0.1. Period 1 refers to the period
September 2005-August 2008. Period 2 covers September 2008-August 2008. Period 3 extends unti l
January 2010. Period 4 las t then unti l August 2012. Period 5 extends unti l January 2014.
Periphera l economies included are Portugal , Ireland, Spain and Ita ly. Core countries in the
sample include Germany, France, Austria , Bel igum and The Netherlands .
Macro factors Financia l Dominance? Including Bank Risk Contagion & Global Dynamic Panel - GMM
Publ ic Debt (% GDP) 2.85074*** 3.03412*** 4.36466** 2.41267 -0.16140
Robust s tandard errors in brackets . *** p<0.01, ** p<0.05, * p<0.1. Banks Home bias refers to asset that are of a domestic nature. Banks private assets refers
to assets not related to the Publ ic sector.Al l bank balance sheet variables are measured as a % of banks ' tota l assets but Bank Assets to depos i ts that
presents the ratio of tota l assets to depos i t l iabi l i ties . Al l explanatory variables enter in the regress ion in lagged form.
Bank factors - ECB dataBank factors - ECB & IMF
dataBank & macro factors Including Sovereign Risk Contagion& Global Dynamic Panel - GMM
Bank Home Bias 645.29571*** -716.28140*** -593.35841*** -603.49704*** -632.80719*** -98.82991***
Robust s tandard errors in brackets . *** p<0.01, ** p<0.05, * p<0.1. Banks Home bias refers to asset that are of a domestic nature. Banks private assets refers to assets not related to the Publ ic sector.Al l
bank balance sheet variables are measured as a % of banks ' tota l assets but Bank Assets to depos i ts that presents the ratio of tota l assets to depos i t l iabi l i ties . Al l explanatory variables enter in the
regress ion in lagged form.
Dep. variable: Sovereign Risk Bank SizeNon-performing
loans
Bank foreign
Liabi l i ties
Bank Risk -0.06955*** -0.06905*** -0.06800***
[0.01003] [0.01025] [0.01041]
Bank Risk* Banks ' Size 0.00071***
[0.00012]
Bank Risk* Non-performing
loans 0.00890***
[0.00197]
Bank Risk*Banks ' Foreign
Liabi l i ties0.01927***
[0.00287]
Constant 81.07124*** 79.63427** 80.33793**
[31.02581] [31.08578] [31.29246]
Other controls Yes Yes Yes
Observations 534 534 534
Number of countries 9 9 9
Robust s tandard errors in brackets . *** p<0.01, ** p<0.05. Other controls include
a l l the regressors presented in the last column of Table 3. Al l the variables
interacted with the SovereignRrisk index are measured as % of GDP.
Table 5. Channels of transmission of Bank Risk
Dep. Variable: Bank Risk Public debtExposure to the
sovereign
Investment grade
effect
Sovereign Risk 0.05698 0.05916 0.07102*
[0.03960] [0.04105] [0.04270]
Sovereign Risk* Bank's
exposure to the Sovereign6.66832***
[2.03944]
Sovereign Risk* Publ ic Debt 0.00380***
[0.00112]
Sovereign Risk* Non-
Investment Grade Dummy0.33462*
[0.17285]
Constant 49.58306 47.43372 48.26995
[41.72077] [43.26140] [46.91890]
Other controls Yes Yes Yes
Observations 534 534 534
Number of countries 9 9 9
Table 6. Channels of transmission of Sovereign Risk
Robust s tandard errors in brackets . *** p<0.01, ** p<0.05. Other controls include a l l the regressors
presented in Table 4. Publ ic debt i s measured as % of GDP. Banks ' exposure to the sovereign is
measured as % of tota l assets .
Bank Risk -0.01432*** -0.01259*** -0.01180*** -0.01273***
[0.00319] [0.00258] [0.00275] [0.00247]
Bank Risk* Bai lout Size
(including contingent cla ims)0.03402***
[0.00101]
Bank Risk* Bai lout Size 0.22592***
[0.02244]
Bank Risk*Bai lout Sizes*Banks '
Foreign Liabi l i ties0.06211***
[0.00365]
Bank Risk*Bai lout Size*Banks '
sovereign exposure4.46927***
[0.43230]
Constant -1.04590 -1.68672 12.64554*** -1.83684
[3.80581] [3.47236] [3.56339] [3.41779]
Other controls Yes Yes Yes Yes
Observations 534 534 534 534
Number of countries 9 9 9 9
Table 7. Bank bailouts and feedback loops
Robust s tandard errors in brackets . *** p<0.01, ** p<0.05. Other controls include a l l the regressors presented in the last
column of Table 3. The bai l out variables are in % of GDP. Banks ' foreign l iabi l i ties i s measured as % of GDP. Banks '
sovereign exposure is measured in % of tota l assets .
Dep. variable: Sovereign Risk
6a Circuit de la Foire InternationaleL-1347 Luxembourg
